BEGIN:VCALENDAR VERSION:2.0 PRODID:-//HasGeek//NONSGML Funnel//EN DESCRIPTION:AI agents don't behave like typical services: they have variab le control flow\, chained model and tool calls\, unpredictable latency and cost. X-WR-CALDESC:AI agents don't behave like typical services: they have varia ble control flow\, chained model and tool calls\, unpredictable latency an d cost. NAME:Platform engineering for production agents - a hands-on workshop X-WR-CALNAME:Platform engineering for production agents - a hands-on works hop REFRESH-INTERVAL;VALUE=DURATION:PT12H SUMMARY:Platform engineering for production agents - a hands-on workshop TIMEZONE-ID:Asia/Kolkata X-PUBLISHED-TTL:PT12H X-WR-TIMEZONE:Asia/Kolkata BEGIN:VEVENT SUMMARY:Platform engineering for production agents - a hands-on workshop DTSTART:20260919T083000Z DTEND:20260919T123000Z DTSTAMP:20260915T101154Z UID:session/WePQpubSxTVsdaDGsuxqe4@hasgeek.com SEQUENCE:13 CREATED:20260828T095012Z DESCRIPTION:# Platform Engineering for production agents\n\n**Check out th e set-up required for the workshop\, and other details from this document - https://docs.google.com/document/d/1prbu5lZoCIduQekNM3Jln0IZbRJcg_d-gzYE 3jhUqDg/edit?tab=t.0**\n \nA hands-on workshop for platform\, systems\, De vOps\, and SRE engineers who need to run AI agents safely\, reliably\, and cost-effectively in production.\n\nAI agents are no longer just local cod ing assistants or experimental chatbots. Once agents call tools\, access d ata\, run workflows\, retry failed steps\, and serve users\, they become p roduction workloads. They need the same platform discipline as any other d istributed system: identity\, secrets\, isolation\, observability\, deploy ment\, rollback\, cost controls\, and operational recovery.\n\nThis worksh op focuses on the runtime and platform responsibilities that come with sup porting agent-based applications in production.\n\n# What is this workshop about?\nThis workshop will help engineers understand what changes when AI agents move from development environments into hosted\, production system s.\n\nParticipants will learn how an agent request flows across model call s\, tool calls\, retries\, responses\, logs\, traces\, and failure states. The workshop will cover the operational controls needed to make agents ob servable\, bounded\, secure\, debuggable\, and recoverable.\n\nThe emphasi s is not on prompt engineering or chatbot design. The workshop is about th e production environment around agents: how to run them\, monitor them\, c onstrain them\, and recover them when something goes wrong.\n\n# Who is th e target audience?\nThis workshop is for:\n- Platform engineers supporting application teams that are building or deploying AI workloads\n- Systems engineers working on infrastructure\, runtime environments\, internal plat forms\, or developer platforms\n- DevOps and SRE practitioners responsible for reliability\, deployment\, observability\, security\, and cost contro ls\n- Backend engineers building agents who want to understand the product ion platform their agents will need\n- Engineers experimenting with AI age nts who understand containerisation basics and want to move beyond local d emos\n\n# Background knowledge requirements\n* _You do not need prior expe rience building AI agents._\n* You should be comfortable with basic contai nerization concepts and familiar with how production applications are usua lly deployed\, monitored\, and operated.\n\nThis workshop is **not** desig ned for non-technical participants\, business users\, or people looking on ly for an introduction to prompting or AI tools.\n\n# Workshop content and plan\nThe workshop will be hands-on and will cover the following areas:\n \n## 1. What makes agents different as production workloads\n- Difference between coding assistants\, local agents\, and hosted agent services\n- An atomy of an agent request\n- Model calls\, tool calls\, retries\, memory\, state\, and responses\n- Why agent workloads are harder to reason about t han standard request-response services\n\n## 2. Runtime behaviour and fail ure modes\n- Tracing an agent request across model and tool boundaries\n- Latency\, timeout\, retry\, and partial failure patterns\n- Model failures \, tool failures\, permission failures\, and workflow failures\n- Choosing safe recovery actions when agent execution goes wrong\n\n## 3. Observabil ity for agents\n- Logs\, metrics\, and traces for agent systems\n- Trackin g token usage\, latency\, errors\, tool calls\, and cost\n- Debugging agen t behaviour using telemetry\n- Separating application bugs\, model issues\ , and platform failures\n\n## 4. Access\, isolation\, and safety boundarie s\n- Bounding what an agent can access\n- Tool permissions and scoped cred entials\n- Secrets management for agent workloads\n- Sandboxing\, isolatio n\, rate limits\, time limits\, and spend limits\n\n## 5. Platform respons ibilities\n- What belongs in agent code vs what should be provided by the platform\n- Minimum production environment for hosted agents\n- Deployment \, rollback\, configuration\, and environment management\n- Operational ch ecklists for supporting agent-based applications\n\n## 6. Hands-on exercis es\nParticipants will work with a small agent workload and practice:\n\n- Running it in a containerised environment\n- Tracing a request through mod el calls and tool calls\n- Observing latency\, failures\, token usage\, an d cost\n- Applying runtime limits and access boundaries\n- Diagnosing fail ures and selecting recovery actions\n\n# Key takeaways\nBy the end of the workshop\, you will be able to:\n\n- Explain how production agents differ from local coding tools and simple AI demos\n- Trace one agent request acr oss model calls\, tool calls\, retries\, and responses\n- Use metrics and traces to identify latency\, errors\, token usage\, and cost\n- Bound an a gent’s access to tools\, data\, time\, credentials\, and spending\n- Dec ide which controls belong in agent code and which should be platform defau lts\n- Define the minimum production environment for identity\, secrets\, isolation\, telemetry\, deployment\, and rollback\n- Diagnose model\, tool \, and agent failures and choose a safe recovery action\n- Support applica tion teams that are moving agent workloads into production\n\n# Software i nstallation requirements\nParticipants will need:\n- A laptop that can run Docker containers\n- An API key for an LLM provider or model gateway\n- A code editor\n- Basic familiarity with command-line usage\n\nYou may use a ny LLM API provider that is compatible with the workshop exercises.\n\nLow -cost or free options include:\n\n- OpenRouter free models: https://openro uter.ai/openrouter/free\n- Command Code Go: https://commandcode.ai/pricing \n- OpenCode Go: https://opencode.ai/go\n\nDetailed setup instructions wil l be shared with registered participants before the workshop.\n\n# About t he instructor\nGiridharan Manivannan is a DevOps and Platform Engineering consultant with over 12 years of experience. He has worked with infrastruc ture teams at Razorpay and Hasura\, helping teams build\, operate\, and sc ale production systems.\n\nHe writes and shares thoughts on DevOps\, platf orm engineering\, AI infrastructure\, Chelsea FC\, and anime at https://x. com/_dforce. He can be reached at hello@giridharan.dev for work.\n\n# How to attend this workshop\nThis workshop is open to [Rootconf members](hasge ek.com/rootconf#memberships).\n\nThe workshop is open to 30 in-person part icipants\, with hybrid access available for remote attendees. Seats for in -person participants will be available on a first-come\, first-served basi s.\n\n# Contact information\nFor inquiries about the workshop\, contact +9 1-7676332020 or write to info@hasgeek.com. LAST-MODIFIED:20260911T060300Z LOCATION:Bangalore - https://hasgeek.com/rootconf/platform-engineering-for -production-agents/ ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com URL:https://hasgeek.com/rootconf/platform-engineering-for-production-agent s/ BEGIN:VALARM ACTION:display DESCRIPTION:Platform engineering for production agents - a hands-on worksh op in 5 minutes TRIGGER:-PT5M END:VALARM END:VEVENT END:VCALENDAR